| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 10
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello!
Maybe there is chance to start some GPU tasks for TB project ? |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
Hello! Maybe there is chance to start some GPU tasks for TB project ? It's up to the project scientists if they want to port the project over to run on GPUs. My guess is it's probably not going to happen.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
![]() ![]() |
||
|
|
Mumak
Senior Cruncher Joined: Dec 7, 2012 Post Count: 477 Status: Offline Project Badges:
|
GROMACS (the app used for HST1) supports GPU acceleration (both CUDA and OpenCL) of some parts of the code.
----------------------------------------If you wish to use the excellent native GPU support in GROMACS, NVIDIA's CUDA version 4.0 software development kit is required, and the latest version is strongly encouraged. NVIDIA GPUs with at least NVIDIA compute capability 2.0 are required, e.g. Fermi or Kepler cards. You are strongly recommended to get the latest CUDA version and driver supported by your hardware, but beware of possible performance regressions in newer CUDA versions on older hardware. Note that while some CUDA compilers (nvcc) might not officially support recent versions of gcc as the back-end compiler, we still recommend that you at least use a gcc version recent enough to get the best SIMD support for your CPU, since GROMACS always runs some code on the CPU. It is most reliable to use the same C++ compiler version for GROMACS code as used as the back-end compiler for nvcc, but it could be faster to mix compiler versions to suit particular contexts. To make it possible to use other accelerators, GROMACS also includes OpenCL support. The current version is recommended for use with GCN- based AMD GPUs. It does work with NVIDIA GPUs, but using the latest NVIDIA driver (which includes the NVIDIA OpenCL runtime) is recommended, and please see the known limitations in the GROMACS user guide. The minimum OpenCL version required is 1.1. ![]() [Edit 2 times, last edit by Mumak at Mar 25, 2016 10:41:41 AM] |
||
|
|
KLiK
Master Cruncher Croatia Joined: Nov 13, 2006 Post Count: 3108 Status: Offline Project Badges:
|
Great, so we might see some real time science on crunching HST research!
----------------------------------------But I'd like to propose admins & techs a setup which will get us option to crunch CPU only, GPU only or CPU + GPU WUs tasks in same time (last one doesn't mean CPU feeder task, but a full CPU task)...can we get that setup in WCG?! ![]() EDIT: Also suggest for "time crunching" in GPU apps use GPU time for time on project...as a GPU is used as a "extra full time core"! If you don't know how it can be done, ask SETi@home guys, they'll help you out in both of these suggestions! ![]() ---------------------------------------- [Edit 1 times, last edit by KLiK at Mar 25, 2016 11:13:02 AM] |
||
|
|
Falconet
Master Cruncher Portugal Joined: Mar 9, 2009 Post Count: 3315 Status: Offline Project Badges:
|
On Facebook, when asked about HST running on GPU's, WCG says:
----------------------------------------"Help Stop TB isn't a GPU-enabled project, but we hope to have another one soon." :D :D :D :D ![]() - AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W - AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W - AMD Ryzen 7 7730U 8C/16T 3.0 GHz |
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
In a year, far far behind us, there was some talk of MCM being GPGPU enabled. The guys in Canada have experience through their HCC1 work.
|
||
|
|
KLiK
Master Cruncher Croatia Joined: Nov 13, 2006 Post Count: 3108 Status: Offline Project Badges:
|
In a year, far far behind us, there was some talk of MCM being GPGPU enabled. The guys in Canada have experience through their HCC1 work. yes, & they came with solution: "making a GPU app wil take us 6m with BETAs & everything it will be over in 9m...while the MCM will be over within a year...so no GPUs for MCM! HST is a different beast...it starts NOW, so it will have some time to develop a GPU app & speed up the process! EDIT: unless, from SekeRob charts, we can read that in 2d it did 1/3 of %...so to 100% it has 300x more...or 600d, so it might end sooner than GPU is available! ![]() ---------------------------------------- [Edit 1 times, last edit by KLiK at Mar 26, 2016 12:10:35 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
EDIT: unless, from SekeRob charts, we can read that in 2d it did 1/3 of %...so to 100% it has 300x more...or 600d, so it might end sooner than GPU is available! ![]() What?????? That's extreamly fuzzy math.... |
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
Just ignore him, which I do permanently,. Did explain in the crunching chart thread how it got there and the fat thumb base (just plugged 2 years in to get a sense of progress on the bar)
|
||
|
|
KLiK
Master Cruncher Croatia Joined: Nov 13, 2006 Post Count: 3108 Status: Offline Project Badges:
|
EDIT: unless, from SekeRob charts, we can read that in 2d it did 1/3 of %...so to 100% it has 300x more...or 600d, so it might end sooner than GPU is available! ![]() What?????? That's extreamly fuzzy math....check the inputs here: https://secure.worldcommunitygrid.org/forums/wcg/viewthread_thread,38947 to do the math after 3d, which says 0,45% in a 3d...gives you around 667d of finishing a work...do that would be Jan 2018! HST is not going to be a LONG project on WCG... ![]() |
||
|
|
|